signalHacker News Best2026-09-26
Ollaya: Ollama for open-source, Jev-style decision models
Ollaya runs open-source decision models locally on your own hardware, answering typed questions in 8-10 ms on an RTX 4090, versus 236-276 ms for TypeSafe Jev's hosted API. It is drop-in compatible with TypeSafe's API, so the official Python SDK 0.7.1 works unchanged. Models like laya (fastest), decider (most accurate), and von (8k context) are available with open weights, and no per-token fees.
- for who
- Developers and teams handling sensitive data who need fast, private decision models.
- why now
- New local, drop-in alternative to TypeSafe's hosted API, eliminating per-token fees while keeping data private.
- what changes
- Decision models run locally on your hardware, keeping data private and eliminating ongoing API costs.
- to do
- Download Ollaya, point the TypeSafe SDK to a local server, and run decision models on your own GPU.
key points
- End-to-end response in 8-10 ms on RTX 4090 for five questions
- Drop-in compatible with TypeSafe API, official Python SDK 0.7.1 works unchanged
- Open models: laya fastest, decider most accurate, von reads 8k tokens
#local deployment#decision models#open source
score
score 8 out of 10. 0-10: how dense the facts are, multiplied by how much you can do with them after reading. 8+ means the topic's evidence bar is met: benchmarks and availability for a new model, amount and investors for a funding round, revenue figures for a solo-money story. Below 5 an item does not enter the digest. A press release scores 3 or less, a reprint loses 2, anything older than 14 days loses 1, a headline that misleads loses 3.
read the source